Human health and the health of ecosystems are inextricably linked. The deteriorating state of the environment influences the health of the human population, which is reflected in the growing number of deaths. According to the new Report No 9/2020 from the EU environment agency (EEA), already one in every eight deaths in Europe can be linked to pollution [1]. Hazardous xenobiotics are usually recalcitrant to degradation and, due to the long-range transboundary migration, can be accumulated in the environment far from the sources of their emission [2]. The scale of the problem was presented in a comprehensive review by Bartrons et al. [3]. Xenobiotics posing a public health hazard are a very diverse group of functional substances, e.g., pharmaceutical compounds, personal care products, pesticides, polycyclic aromatic hydrocarbons (PAHs), or textile dyes. Unfortunately, most of them represent persistent organic pollutants (POPs) and, despite the great removal effort, many of these substances persist in the environment as micropollutants affecting human health. The main source of their intake is inhalation with air or consumption of contaminated edible plants and meat coming from animals fed with polluted crops.

Plants may become contaminated with organic pollutants both as a result of the deposition of these compounds along with dust on the surface of leaves and through the uptake thereof with roots from the soil. The method of uptake and translocation of particular organic compounds into plant tissues depends mainly on their physical and chemical properties, e.g., molecular mass, hydrophobic/hydrophilic properties expressed as octanol—water partition coefficients (K_OW), biological characteristics of the plant, and soil features ^[4][5][6][4,5,6]. Although the solubility and concentration of organic pollutants in the soil have a lower impact on the above-mentioned processes, they should be taken into account as well [5]. Globally, the principal source of POPs in vegetation in remote and rural areas is the atmosphere, while their acquisition from soil to plant roots is a secondary entrance pathway [3]. In turn, in industrialized and urbanized areas, the main source of organic contaminations in plants is preferentially the soil.

Moderate hydrophilic substances with high solubility in water and a simultaneous ability to permeate lipid membranes are easily absorbed by plant roots and translocated to different aerial parts. The penetration of hydrophobic pollutants from the soil into plants is more complex due to their weak solubility and low bioavailability. Thanks to the active desorption mechanism with the participation of plant root exudates and binding proteins, hydrophobic contaminants are released from soil particles and can be uptaken. Since the translocation thereof is a structure-related process and takes place only for a few of them, the highest accumulation of soil hydrophobic pollutants is observed in lipophilic tissues of roots ^[3][4][5][3,4,5]. After the uptake by plants, organic pollutants may be metabolized and/or accumulated inside plant tissues or volatilized into the atmosphere. The efficiency of these processes can be enhanced with the help of plant-associated bacteria, that have the ability to transform such substances through metabolic or enzymatic processes: growth and co-metabolism, respectively [7]. Plant-associated bacteria include endophytic, phyllospheric, and rhizospheric bacteria. Endophytes seem to be the best choice for the improvement of phytoremediation.

Endophytes colonize tissues of the living plant without expressing any visible signs or symptoms. Although successful endophytic colonization is dependent on many factors, such as the host species, plant organs, geographic locality, or seasonality, when already established it is more stable than the interaction of rhizospheric bacteria with plants. Thanks to the close contact with plant cells, endophytes are able to communicate and interact with the plant more efficiently. Additionally, they do not need to compete for nutrients and the niche with the dense population of autochthonous or indigenous bacteria in the rhizosphere, and they are protected from extreme abiotic conditions. These features allow maintenance of their high abundance, which is essential for the degradation of pollutants. Moreover, unlike rhizospheric bacteria, besides the direct reduction of the content of xenobiotics inside plants and in their environment through many mechanisms, endophytes simultaneously stimulate plant defense mechanisms. Both routes counteract with abiotic stress induced by pollutions in plants. Moreover, genomic studies of endophytes have demonstrated that these microbes are far more versatile than rhizospheric bacteria and may contain genes for novel traits that are beneficial to the host plant, among them degradative ones ^[8][9][10][8,9,10]. Besides, endophytic bacteria can regulate the metabolic processes of organic contaminants in plants through horizontal gene transfer to native endophytes or to the host and gene duplication ^[11][12][11,12].

Since it has been suggested that the endophyte microbiome may be a subpopulation of rhizosphere-inhabiting bacteria [8], some attempts have been made to identify genomic markers of the endophytic lifestyle. Genome comparisons between bacterial endophytes and the genomes of rhizospheric plant growth-promoting bacteria indicated no definitive group of genes responsible for the colonization of plants; however, these studies are starting to unveil potential genetic factors involved in the endophytic lifestyle ^[8][13][8,13]. Among the bacterial genes expressed in planta and allowing colonization, two genes involved in alkane degradation are mentioned: alkB and CYP153 [14].

2. Evidence of Benefits from the Plant-Endophyte Partnership in Proximity of Xenobiotics

Advanced treatment processes are necessary for the effective removal of organic pollutants. Some of these methods are ozonation, ultrasound, ultraviolet, Fenton processes, membrane systems, biosorption, and biodegradation both in situ and ex situ depending on environmental matrices to be treated. However, recent reports have suggested that more than one treatment technique may be required to degrade these compounds completely [15]. Thus, synergistic interactions between plants and interior plant tissue bacteria seem to be a promising approach for the effective removal of residual recalcitrant organic compounds. The first attempts to prove the validity of this approach have already been carried out, and endophytic bacteria with the potential to be used in microbe-assisted phytoremediation have been mostly acquired from plants grown on contaminated soils ^[7][16][17][7,16,17]. Some strains able to colonize plant tissues and degrade xenobiotics were also obtained from contaminated sediments and soils ^[18][19][18,19] and, what is less obvious, from plants grown on non-contaminated sites ^[20][21][20,21]. The most commonly isolated bacterial endophytes from those niches were assigned to the genera Pseudomonas, Bacillus, Burkholderia, Stenotrophomonas, Micrococcus, Pantoea, and Microbacterium. They were shown to have versatile metabolic pathways for utilization of organic pollutants as the only source of carbon but more frequently and efficiently in co-metabolism, which consequently enables the microorganisms to mineralize or transform contaminants into non-toxic derivatives. However, in order to remove contaminants effectively, partners must act synergistically. The first crucial step of the degradation of anthropogenic organic pollutants inside plants consists in the activation of aromatic rings with the participation of bacterial endophyte oxygenases followed by the action of other enzymes, e.g., esterases, reductases, or dehalogenases. In contrast, plants can increase the efficiency of the degradation by providing the bacterial partner with additional sources of carbon and nitrogen [7].

2.1. Removal of Hydrocarbons

Hydrocarbons comprise a broad family of aliphatic, aromatic, and polycyclic compounds with high carbon ranges. They are ubiquitous environmental pollutants generated primarily from oil spillage, pesticides, automobile oils, urban stormwater discharges, and other anthropogenic activities; nevertheless, some originate from natural sources. In some national and international documents related to risk assessment for both ecological and human exposure to petroleum hydrocarbons (PHC), the assumption that plants are unable to take up petroleum hydrocarbons from contaminated soil has appeared and, therefore, subsequent exposure at higher trophic levels is not a concern [22]. However, various studies based on chemical analyses suggest that plants are not only able to absorb PHC into their tissues, but that there is a noticeable upward trend in the hydrocarbon concentrations of the vegetation over time ^[3][22][3,22]. Since they are highly lipid-soluble and can be readily absorbed from the gastrointestinal tract of mammals and many of them have toxic, mutagenic, and/or carcinogenic properties, there is an urgent need to develop safe and efficient ways for removal or degradation of these contaminants [23]. It has been shown that organic contamination of soil may affect the population characteristics of endophytic bacteria [24]. For instance, in their study on bacterial community in ryegrass (Lolium multiflorum Lam) exposed to phenanthrene and pyrene in comparison to non-contaminated plants, Zhu et al. [25] showed that strains from the genera Bacillus, Pantoea, Pseudomonas, Arthrobacter, Pedobacter, and Delftia were present only in plants exposed to PAHs. This may suggest their potential for biodegradation of the hydrocarbons tested. Moreover, it was shown that the higher concentrations of individual or combined PAHs were accompanied by lower biodiversity of endophytes [25]. In turn, it was found in another study that inoculation of phenanthrene-contaminated wheat with PAH-degrading endophytic Massilia sp. Pn2 had an impact on the endophytic bacterial community structure: diversity and richness as well as the overall bacterial cell counts. Also, in this case, these relationships were associated in a contamination level-dependent manner [26]. These and similar findings may indicate the direction of further research. Although a variety of hydrocarbon-degrading plant-associated bacteria has been isolated and characterized till now, only some of them were proved to exhibit an endophytic lifestyle. The first studies on bacterial endophytes were focused on their suitability to degrade hydrocarbons in in vitro cultures and decontaminate polluted soils. In experiments conducted by Pawlik et al. [24], more than 90% of isolates obtained from Lotus corniculatus L. and Oenothera biennis L. grown in long-term PHC-polluted sites and classified to the genera Rhizobium, Pseudomonas, Stenotrophomonas, and Rhodococcus were confirmed to be able to utilize diesel oil as a carbon source. Also, Pseudomonas aeruginosa L10 isolated from the roots of a reed Phragmites australis was shown to participate in degradation C₁₀-C₂₆ n-alkanes in diesel oil, as well as naphthalene, phenanthrene, and pyrene in individually enriched cultures. Furthermore, L10 was able to increase the petroleum hydrocarbons (PHCs) degradation rate in pot trials. These findings were confirmed by genome annotation, which indicated the presence of genes related to the n-alkane and aromatic compound degradation pathways in L10 [27]. The colonization of plant tissues by endophytic strains potentially involved in hydrocarbons degradation was confirmed by many other authors also with the use of PCR amplification of the following alkane-degradation genes: alkH (alkane hydroxylase), alkB (alkane monooxygenase), c23o (catechol-2,3-dioxygenase), CYP153 (cytochrome P450-type alkane hydroxylase) and aromatic compound pathway genes: pah (alpha subunit of the PAH-ring hydroxylating dioxygenases) or ndoB (naphthalene dioxygenase) ^{[16][24][28][29]}[16,24,28,29]. The presence of such genes was most commonly found in strains classified to Bacillus and Pseudomonas and less frequently detected in Microbacterium, Rhodococcus, Curtobacterium, Pantoea, and Enterobacter ^[14][28][29][14,28,29]. Compared to classical phytoremediation, the higher benefits of cooperation of endophytic strains with their host plants were observed as a higher decrease in the content of pyrene, anthracene, PHCs, or PAHs in the soil was established for Stenotrophomonas sp. EA1-17, Flavobacterium sp. EA2-30, Pantoea sp. EA4-40, Pseudomonas sp. EA6-5, Enterobacter sp. 12J1, Enterobacter ludwigii ISI10-3 and BRI10-9, Bacillus sp. SBER3, Bacillus safensis ZY16, and Burkholderia fungorum DBT1 ^{[17][18][19][20][29][30]}[17,18,19,20,29,30]. In a similar approach, the possibility of degradation of a mixture of PAHs (naphthalene, phenanthrene, pyrene, fluoranthene) with high concentrations by endophytic Stenotrophomonas sp. P₁ and Pseudomonas sp. P₃ isolated from tissues of Conyza canadensis and Trifolium pratense L., respectively, was demonstrated [7]. In turn, Paenibacillus sp. PHE-3 isolated from Plantago asiatica L. exhibited an ability to degrade HMW-PAHs in the presence of other 2-, 3-ringed PAHs through co-metabolism [31].

2.2. Decontamination of Textile Dyes

The use of dyes in textile, leather, cosmetic, pharmaceutical, and paper industries is one of the most environmentally polluting and devastating anthropogenic activities, additionally posing health hazards to humans ^[32][44]. Since they are usually water-soluble organic compounds, which can penetrate plant and animal tissues ^[33][45], the effective discharge of hazardous dyes from aqueous solutions and detoxification is crucial. There are various physical, chemical, and biological methods available for the removal of dyes from wastewater, but phytoremediation is generally considered to be the most promising and low-cost approach. Although plants play a significant role in the direct uptake of pollutants from wastewaters, the processes of transformation and mineralization of textile dyes greatly depend on microbial communities closely associated with their roots systems. Different endophytes can decontaminate textile dye wastewater through bioaccumulation, biosorption, or biotransformation, which results in not only decolorization but also detoxification of dyes in the environment. Thus, the biodegradation of textile dyes by the synergistic action of endophytes and plants seems to be a viable alternative to pure classical phytoremediation. Textile dyes can be classified into many groups based on the structure of the chromophore. However, the most prevalent are azo dyes, anthraquinones, and triphenylmethanes. Among them, azo dyes are common xenobiotic and recalcitrant materials, due to the high stability of the azo groups (–N=N–). In order to decolorize azo dyes, it is necessary to break double chromophore bonds, but since they are very stable, their degradation with conventional physicochemical methods is usually not possible ^[34][46]. Anthraquinone dyes are the second largest class of dyes containing a fused aromatic ring structure, which makes them recalcitrant to degradation. These dyes are characterized by the presence of the chromophore group =C=O. Among triphenylmethane, crystal violet had the most stable structure due to the presence of the quaternary ammonium substituent ^[35][47]. According to the selection rule, endophytes isolated from plants growing in contaminated areas should be able to biodegrade various dyes. For example, Exiguobacterium profundum strain N4 obtained from Amaranthus spinosus collected from a site polluted with effluents from textile dyeing and printing industries was able to bleach and degrade diazo dye Reactive Black-5 by enzymatic oxidation, reduction, desulfonation, and demethylation to nontoxic benzene and naphthalene [9]. Similarly, the alkaliphilic endophyte Bacillus fermus (Kx898362) obtained from Centella asiatica showed the potential to degrade diazo dye Direct Blue-14 in in vitro assays. The disintegration patterns revealed by LC-MS showed that the parent DB-14 molecule was completely disintegrated into five noncytotoxic intermediates ^[34][46]. In turn, the endophytic bacterium Klebsiella aerogenes S27 obtained from the leaves of the wetland plant Suaeda salsa was involved in the biodegradation of triphenylmethane dye malachite green (MG) into a nontoxic metabolite N,N-dimethylaniline. The removal of MG is of great importance, since it had been extensively used in dye industries or in aquaculture as an antifungal agent before 1993 when it was nominated as a priority chemical for carcinogenicity testing by the United States Food and Drug Administration (FDA) ^[33][45]. The inoculation of PGP-endophytes to plants growing in soil irrigated with textile effluents for improvement of plant biomass production and for soil remediation is still a rare practice. Several reports are available in the literature on the bioremediation of dyes by endophytic microorganisms, mostly used in phytodepuration systems. Spectrometric analysis of the end products of degradation of sulfonated diazo dye Direct Red 5B showed that the synergistic action of the Portulaca grandiflora plant and Pseudomonas putida strain PgH resulted in higher biotransformation with enhanced efficiency than when each of them acted separately. Moreover, a phytotoxicity study revealed the non-toxic nature of metabolites formed after parent dye degradation ^[36][48]. Also, the collective action of endophytic Microbacterium arborescens TYSI04 isolated from shoots of Typha domingensis and Bacillus pumilus PIRI30 obtained from roots of Pistia enhanced textile effluent degradation and toxicity reduction, which was confirmed by significant reductions in chemical oxygen demand—COD (79%), biological oxygen demand—BOD (77%), total dissolved solids—TDS (59%), TSS (27%), and color removal within 72 h when a combination of plants and bacteria was applied ^[37][49]. A similar effect was achieved by Nawaz et al. ^[38][50] with the use of a consortium consisting of PGP strains (i.e., Acinetobacter junii NT-15, Rhodococcus sp. NT-39, endophytic Pseudomonas indoloxydans NT-38), and Phragmites australis for removal of three commonly used acid metal textile dyes containing two sulfo groups: Bemaplex Navy Blue D-RD, Rubine D-B, and Black D-RKP Bezma from water. Based on in vitro and in vivo characterization, in terms of Reactive Black 5 decolorization activity, a consortium of strains Pseudomonas fluorescens CWMP-8R25, Microbacterium oxydans CWMP-8R34, Microbacterium maritypicum CWMP-8R67, Flavobacterium johnsoniae CWMP-8R71, Lysinibacillus fusiformis CWMP-8R75, and Enterobacter ludwigii CWMP-8R78 isolated from P. australis was identified as promising in phytodepuration systems. The F. johnsoniae and E. ludwigii strains also decolorized Bezactive rouge S-Matrix, Tubantin blue, and Blue S-2G in an in vitro assay ^[39][51]. Some endophytic bacteria possess biosorption and bioaccumulation properties that can be exploited in dye decontamination. However, the bioaccumulation process is usually not preferred, compared to biosorption, because the live microbial biomass requires nutrients and supplements for its metabolic activities, which in turn would increase BOD or COD in the aquatic environment. In turn, biosorption via mechanisms such as adsorption, absorption, ion exchange, precipitation, and surface complexation needs large amounts of biomass, which is economically and technologically unfavorable. Thus, the main mechanism of biotransformation of pollutant dyes by bacterial endophytes takes place through the action of highly oxidative and non-specific ligninolytic enzymes: laccase, azo reductase, peroxidases, tyrosinase, and hydrogenase ^[40][52]. In the azoreductase-mediated cleavage of the azo bonds, toxic aromatic amines are released, which next need to be transformed into non-toxic compounds. Moreover, azo reductases are oxygen-sensitive and degrade azo dyes only in the presence of reducing equivalents FADH and NADH anaerobically. Unlike peroxidases, laccases oxidizing a wide range of polyphenols, methoxy-substituted phenols, and diamines do not produce toxic peroxide intermediates from azo dyes. Bioinformatic analysis carried out by Ausek et al. ^[41][53] revealed a high diversity of genes for laccase-like enzymes among diverse bacteria, including the most common endophytic genera Streptomycetes, Bacilli, and Pseudomonads as well as anaerobes, autotrophs, and alkaliphiles. Additionally, most of them had signal peptides indicating that these laccases may be exported from the cytoplasm, which improves their potential for future biotechnological application. However, only a few of them were detected in strains obtained from plant tissues. It was shown that the endophytic bacterium Pantoea ananatis Sd-1 isolated from rice seeds produced both intra- and extra-cellular laccases, of which extracellular Lac4 exhibited degradation of non-phenolic and phenolic compounds and decolorization of various synthetic dyes (azo dye Congo Red, anthraquinone dye Remazol Brillant Blue-R, and dyes from the group of triphenylmethane Aniline Blue) ^[42][54]. The laccase gene and activity was also confirmed in the Sinorhizobum meliloti strain L3.8 isolated from root nodules of Medicago sp. ^[43][55]. Another extracellular oxidoreductase enzyme triphenylmethane reductase-like (TMR-like) was involved in the biodegradation of malachite green by endophytic Klebsiella aerogenes S27. Since there is no report on plants harboring tmr genes, bacteria possessing the gene could be very valuable in endophyte-assisted phytoremediation ^[33][45].

2.3. Bioremediation of Polyhalogenated Organic Compounds—Biphenyls and Dibenzodioxins

Polychlorinated biphenyls (PCBs) are classified as POPs with high toxicity. Many of these pollutants, among them bisphenol A (BPA), are recognized as endocrine-disrupting compounds (EDCs) due to their ability to interfere with the human endocrine system. At low concentrations, BPA can also show acute toxicity toward aquatic organisms and carcinogenic properties ^[44][56]. In turn, members of the family of polychlorinated dibenzodioxins (PCDDs) can bioaccumulate in humans and wildlife due to their lipophilic properties and may cause developmental disturbances and cancer. The European Union Water Framework Directive ^[45][57] and the Directive of the European Parliament and Council (2013/39/EU) regarding priority substances in the field of water policy (Directive EQS) list 45 substances representing a serious threat to aquatic environments and to humans, which need to be removed from aquatic environments, including PCBs and PCDDs. Recently, the potential for improvement of removal of BPA in planta has been shown by endophytic Pantoea anantis in combination with its host plant Dracaena sanderiana. Due to the activities of the plants and microorganisms, such physicochemical indicator parameters as pH, COD, BOD, TDS, conductivity, and salinity were reduced after 5 days of the experimental period with a decrease in BPA levels ^[44][46][56,58]. Bioremediation of the most toxic dioxin congener 2,3,7,8-TCDD was shown in a study involving the endophytic bacterium Burkholderia cenocapacia 869T2 isolated from roots of vetiver grass. In an in vitro assay, it was capable of TCDD degradation by nearly 95% after one week of aerobic incubation. Generally, in the bioremediation of dioxins by bacteria, angular dioxygenase, cytochrome P450, lignin peroxidase, and dehalogenases are known as important dioxin-metabolizing enzymes. Through transcriptomic analysis of strain 869T2 exposed to TCDD, a number of catabolic genes involved in dioxin metabolism were detected with high gene expressions in the presence of TCDD. Assays with cloned l-2-haloacid dehalogenase (2-HAD) indicated that it might play a pivotal role in TCDD dehalogenation ^[47][59].